• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

Indie Game Development Discussion Thread | Of Being Professionally Poor

Status
Not open for further replies.

missile

Member
Dude, yes! You in turn got me thinking more about the way I was going to apply my predetermined wind direction/force multipliers to the bird in Unity. The original plan was to use area triggers as my own wind zones, such that when the bird is within its bounds, my basic aerodynamic functions would be influenced by whatever force multipliers I choose for that zone. I was going to manually arrange a number of zones with varying magnitudes so as to "fade" the wind forces when flying between no-, low-, or high-wind zones. ...
Would be cumbersome, indeed.

... I never actually considered it until reading those brief thoughts on convection modeling, but I'm positive I could arrange my wind zones as cubes in a 3D grid, write some functions allowing each zone's wind force/direction values to be influenced by a collective sampling of the force/direction values of the 6 (at most) directly adjacent zones, calculate that influence on a fixed timestep, staggering the calculation between groups of non-adjacent zones (i.e. if it were a 2D grid like a chess board, calculate only white zones, then only black zones, then only white, and so forth), play with methods of handling recursive influence for dampening purposes... essentially creating a very basic fluid model and scale the "resolution" of the 3D grid to the target hardware. Could specify certain zones as "wind sources" by not allowing their force/direction values to be influenced by adjacent zones. I also think it'd be fairly easy to create a few zones representing primitive solids by having them place limits on the wind direction values (within specified ranges) for any adjacent, non-static wind zones. Layer those "solid" zones roughly within corresponding world geometry so that wind zones surrounding a large hill or building, for instance, would be forced to flow around it accordingly. ...
What you've read is essentially the finite difference scheme for computing
solutions of a convective partial differential equation on a structured grid
according to some boundary conditions.

Indeed, that's something one has to consider, at least. But for real wind
there is much more to do. Looking at the 3d Navier-Stokes equation one can see
that it has an advective-convective term, a diffusion term, a pressure term,
a time-derivative, and a force term. Each term must be represented/modeled on
the grid with respect to the unknowns. Looking at all the differential terms,
this will be a very tedious task, and an error prone one if one doesn't pay
minute attention to detail. However, that's just one part of the story. The
other part is the solution method(s) used for the discrete systems of
equations stemming from the finite differencing scheme. Cool physics (non-
linear effects) do often need much better solution methods (implicit methods
etc.) to prevent blow-up. Good iterative solvers are needed as well. Proper
damping does also play a role with respect to convergence of an iterative
solver. But! Too much dumping an the solution looks lame.

I've put myself under development within this regard doing some research and
collecting everything together needed to build a 3d fluid solver. But I won't
start implementing until I've finished some wings (a simple aircraft) for my
rigid body solver. So I think I'm going to start in around one month with all
of the fluid stuff.

... Of course, using heat as a universal upwards force modifier would be great, and I could denote "heat source" zones or groups thereof and base their heat values on any number of calculations like the one you outlined above. ...
The possibilities are endless once a proper model in implemented! :D

Another cool thing is to have (massless) particles transported with the wind.
That way one can actually see the motion and curvature of the wind field.
Given Oculus Rift, this must look stunning!

Anyhow. Basically, I need the wind field for my air-drifting game. With wind
the game will be much more sophisticated and much more fun to play since the
wind may either help you in getting good air or may kick you right off the
play field or onto the ground if you don't know how counteract it. And it
produces sort of an uncertainty. Wind does change, does it? xD
 
I've been messing about in Unity for a couple weeks now and I have to say I'm really liking it, cuts down on "drawing" things and instead lets me focus much more on logic (I'm doing this as a way to keep up on C# programming knowledge while I do SQL work at my job).

My biggest problem so far has been trying to come up with little systems/projects to try and build. There doesn't happen to be a list of good "unity challenges" floating around?
 

omg_mjd

Member
I've been messing about in Unity for a couple weeks now and I have to say I'm really liking it, cuts down on "drawing" things and instead lets me focus much more on logic (I'm doing this as a way to keep up on C# programming knowledge while I do SQL work at my job).

My biggest problem so far has been trying to come up with little systems/projects to try and build. There doesn't happen to be a list of good "unity challenges" floating around?

Not a list, but the next Ludum Dare is in 23 days. Lots of people use Unity so you'll be in good company. Plus, participants are required to share their code so if you see something interesting you can download the project files and see the scripts for yourself.
 

Paz

Member
I know I don't post here as much as I really should, but damn this place can be a font of creativity. You all rock and whenever I see someone looking for interesting indie developments I try and point them here.

That said, nobody has figured out the best solution to accounts based leaderboards on PC :p Shouldn't there be a service for this stuff that's cheap and convenient?
 

beril

Member
ok here goes nothing. first ever gif of my new game

queststory.gif
 

beril

Member
Nice! IIRC, your previous game was made using your own engine? Is that still the case?

Yes same engine, but very very different graphics

It's still early in development. It's been going really slowly lately and I haven't been getting much work done, so there's not much I can show. But I felt I needed to get something out there and get some reactions before I went insane with built up anxiety over it.

One more gif
queststory2.gif
 

bumpkin

Member
Yes same engine, but very very different graphics

It's still early in development. It's been going really slowly lately and I haven't been getting much work done, so there's not much I can show. But I felt I needed to get something out there and get some reactions before I went insane with built up anxiety over it.

One more gif
queststory2.gif
Still, looks pretty freakin' good so far man! When it's ready, I'll be watching for it to hit the appropriate shops with my money ripe for the taking. :)
 

JulianImp

Member
ok here goes nothing. first ever gif of my new game

queststory.gif

Yes same engine, but very very different graphics

It's still early in development. It's been going really slowly lately and I haven't been getting much work done, so there's not much I can show. But I felt I needed to get something out there and get some reactions before I went insane with built up anxiety over it.

One more gif
queststory2.gif

That looks great! I'll be on the lookout for this game. By the way, what platform/s do you intend to release it on?
 

razu

Member
ok here goes nothing. first ever gif of my new game

queststory.gif

Looks cool man :D I'd love to do some 3DS stuff... *sigh*


In other news...

SALES REPORT!!

After putting Chopper Mike back to Tier2, it sold 3 copies!! So, going from free to paid has not increased paid sales!
 

JulianImp

Member
Thanks. I'm focusing on 3DS for now, then I'll see after its done if bringing it to other platforms makes sense

I'm definitely going to get it on 3DS when it's out. Having what seems like a dedicated jump button looks like it could lead to some stuff most Zelda games haven't done before. Best of luck with the development!
 

Ydahs

Member
Looks cool man :D I'd love to do some 3DS stuff... *sigh*


In other news...

SALES REPORT!!

After putting Chopper Mike back to Tier2, it sold 3 copies!! So, going from free to paid has not increased paid sales!
Darn... are you sure that took into account a full day of sales though?
 

missile

Member
Did a quick test of the Rift today.

Contra:
- resolution needs to be improved (1080p already in the works, np)
Pro:
- fast response
- representation of scale is breathtaking!
 
But for real wind there is much more to do.
Oh, of course! I'm just looking to emulate gentle breezes and curvatures in wind direction with a relatively low-resolution grid of zones so that I can play with dynamic wind effects without having to set them all individually, as that would indeed be very cumbersome. I'm happy to settle for linear reactivity among zones as long as it flows around the landscape reasonably, which I believe I can achieve with my solid zones. Not looking for super realistic fluid behavior like generation of turbulence, vortexes, and the like... my zones, acting as individual points of articulation within the "fluid" field, will still be much larger than the bird object I intend for them to affect. Way fewer variables than your intended model, but the end result should be sufficient for my needs. :)

Another cool thing is to have (massless) particles transported with the wind.
.
Definitely. I intend to have an arrow rendered in each zone as I'm prototyping my model, pointing in the direction of flow and fading from red to green as force increases. That should allow me to visualize the reactivity and tune my equations as necessary, after which I could implement some particles to help the player identify wind speed and direction.

I genuinely can't wait to see where you take your engine, especially in terms of the fluid modelling. Thanks for inspiring me to envision a more basic approach for my needs. I'm excited. Too many ideas.
 
Damn man, that sure is shiny! What are you using for an engine?

That is the Sunburn Engine by Synapse Gaming.

I totally see why people would wonder about the music choice. Thexder used the same piece and that was about a transforming robot jet too. Just a nod. The whole trailer was done last night in about 4 hours so I scrambled to get a piece. I'm away on vacation starting tomorrow so had to blitz it out.
 
Hi guys,

Does anyone know (or have links) how to package a game that requires the user to enter a unique product key to install? For now, Windows OS would be ok as a start.


I'm in the process of making my very first game from scratch! It's a puzzle game for starters, while I learn the basics without much collision and moving backgrounds, etc. to get my feet wet. I'm using C++ SDL. Thanks!
 

missile

Member
Oh, of course! I'm just looking to emulate gentle breezes and curvatures in wind direction with a relatively low-resolution grid of zones so that I can play with dynamic wind effects without having to set them all individually, as that would indeed be very cumbersome. I'm happy to settle for linear reactivity among zones as long as it flows around the landscape reasonably, which I believe I can achieve with my solid zones. Not looking for super realistic fluid behavior like generation of turbulence, vortexes, and the like... my zones, acting as individual points of articulation within the "fluid" field, will still be much larger than the bird object I intend for them to affect. Way fewer variables than your intended model, but the end result should be sufficient for my needs. :)
Ahh, understood! Though you found out about finite differences the other way.

Computational resources are demanding in any model which will be a lil more
realistic, indeed. Anyhow, I have some tricks to attack on this problem. Here
is one (the idea); One of my tricks to lower the computational requirements
consist of tweaking the equation solvers while keeping the model as real as
possible. Basically, the convergence rate of many iterative solvers depend on
the goodness of the function in question. My trick here is to artificially
introduce "goodness" and take it away later on. Goodness essentially means
dumping the energy which will accelerate the convergence rate tremendously.
This way a solution can be computed over the entire grid with just a few
iterations, instead of thousands, leading to sort of a damped solution. Now
what? Now I pump some specific amount of energy back into the system at some
specific points within the field such that the solution so obtained isn't so
far away from the one you would get only from very long iterations. Tricky,
heh? ;)

Basically, let the scientist iterate till the end of days to obtain the
solution. We game developers will simply cut-off at the realtime limit and
just throw the missing energy back into the system! *Don't tell any scientist
I ever said that!* Obviously, some physical state equations won't hold
equality this way, but as a gamer you will never going to recognize it. ;)

With such tricks I hope to be able to compute rather cool wind fields with
just a few iterations while the higher physical model is in full effect.

Well, I made an interesting observation;
(a) lower physical model -> solved almost exactly => fast
(b) higher physical model -> solved approximately => fast, as well!

The computational advantage you get while using a lower physical model is
similar to using a higher physical model while solving it by a very
approximative methods. If this holds true, the advantage of (b) becomes pretty
clear. It is much more easy to improve the solver instead of fixing the
physical model for more realism which at times require an entire different
solution/solver approach. And with the increase in computational resources
each year, or considering just different platforms, it becomes petty easy to
utilize the increase in computational resources by just tuning all the
solver's threshold parameters yielding more realism without changing the
physical model at all. So basically, that's way I'm aiming for a more
realistic fluid model here. However, the downside is the upfront cost you have
to pay to even get the higher physical model up and running, which is actually
the advantage of (a).

Edit: I know I have to prove/show this.


... Definitely. I intend to have an arrow rendered in each zone as I'm prototyping my model, pointing in the direction of flow and fading from red to green as force increases. That should allow me to visualize the reactivity and tune my equations as necessary, after which I could implement some particles to help the player identify wind speed and direction.
Yeah. A few hours ago I had an idea to release particles from the wing tips of
an aircraft. Depending on the wind field they will eventually flow at
different velocities and in different directions. Lots of cool stuff incoming,
I guess!

... I genuinely can't wait to see where you take your engine, especially in terms of the fluid modelling. Thanks for inspiring me to envision a more basic approach for my needs. I'm excited. Too many ideas.
Inspiring people always inspires me! Let me see your results. Don't keep me
waiting for too long!

In about a month I'm going to start with the fluid stuff. You may see some
of my progress over here at http://www.nihilogames.com/coding.html where I've
already outlined the direction of attack. During August I'm still keep pushing
hard on the rigid stuff.
 

fin

Member
Way cool! Seems you will have a tough job in generating good levels, do you?
It seems that each puzzle needs to be touched by hand, or can you use some
sort of algorithmically generated puzzle?

Nope, it's all made by hand. It's pretty easy though, I make prefabs out of EVERYTHING!
 

cbox

Member
Showing off the style points feature where you can whack a mine with the whip for extra power :D

444.gif.gifvaa5p.gif


We redid our sparks, and added some extra particles that remain after you kill an enemy thanks to some comments in this thread, improves the look 10 fold. You can even see the "wolfpack" enemy right at the end, they usually stay away from you and actively hide unless there's 11 or more on screen, then they all come screaming at you.

As seen here :

572d3pvx.gif
 

ninjarr

Neo Member
We made a short game for a game-jam that became slightly more. It's a mario parody (not to be confused with mario party) we made in two weeks. Just a short little bout of fun, thought you guys might enjoy. :) I should mention that it currently requires a bit-torrent client to download.
http://www.thewildeternal.com/2013/07/30/sorry-mario-bros/

We made it in Unity as our first 2D project. Our primary project, The Wild Eternal, is in 3D and it was quite nice to play with a different part of my brain developing this. Hope someone enjoys, let me know if you get all 10 peach coins!
 

Dynamite Shikoku

Congratulations, you really deserve it!
Thanks. Yes, $0.99, with one optional $0.99 IAP which unlocks all the levels straight away (otherwise levels just unlock as you play through)
 

Hazaro

relies on auto-aim
Showing off the style points feature where you can whack a mine with the whip for extra power :D

We redid our sparks, and added some extra particles that remain after you kill an enemy thanks to some comments in this thread, improves the look 10 fold. You can even see the "wolfpack" enemy right at the end, they usually stay away from you and actively hide unless there's 11 or more on screen, then they all come screaming at you.

As seen here :

572d3pvx.gif
Visual feedback looks a ton better!

I had trouble identifying the mines at first though. Granted I am not playing, but maybe something like a very fast animated collapsing ring would help clearly identify a placement and give a bonus 'reverb' bit of visual feedback (collapsing circle -> shockwave detonation).
 

missile

Member
Showing off the style points feature where you can whack a mine with the whip for extra power :D

444.gif.gifvaa5p.gif

...
572d3pvx.gif
Pretty cool. But the screen gets cluttered pretty fast with all the icons
on it. What I don't like is the color scheme for all the items. Don't know,
it looks way too passive for me. Some vibrant accents could make the action
look more dynamic/thrilling and may serve for a better discriminability of
some icons -- which at the same time may reduce the visual perception of
clutter to some degree.


Small progress on Steel Storm Forgotten Prison DLC:
...
4ZlVt41.png
What's the game about?
Btw: Welcome over here!
 
Status
Not open for further replies.
Top Bottom